The Peculiar Market for Academic Journals Ted Bergstrom UCSB

Conferring with an industry leader Industry Leaders Conferring with an industry leader

A curious market structure Private profit-maximizing firms and non-profit societies and academic presses both are significant players. Most of the workforce--authors and referees--work for free.

Pricing of Paper Editions In 2004, of the 25 most-cited journals in economics, 18 are owned by non-profit groups. Average price to libraries is $300 per year. Only 7 of the top 25 journals are owned by commercial publishers. Average price to libraries is $1900 per year.

Costs of Economics Journals in 2000 Publisher Type Number of Journals Price per Page Price per Cite Non-Profit 91 $0.18 $0.15 For-Profit 206 $0.82 $2.40

Journal Prices by Discipline (In US $) Cost per page Cost per cite For-profit Non-profit For-profit Non-profit Ecology 1.01 0.19 0.73 0.05 Economics 0.83 0.17 2.33 0.15 Atmosph. Sci 0.95 0.88 0.07 Mathematics 0.70 0.27 1.32 0.28 Neuroscience 0.89 0.10 0.23 0.04 Physics 0.63 0.38

Costs of a Complete Economics Collection Publisher Type Percent of Cost Percent of Cites Non-Profit 9% 62% For-Profit 91% 38%

Costs of a Physics Collection: Heinz Barshall’s List of Journals Publisher Type Percent of Cost Percent of Cites Non-Profit 39% 79% For-Profit 61% 21%

Division of Labor Illustrated by academic journals: The greatest improvements in the productive powers of labour… seem to have been the effects of the division of labour…. Adam Smith, Wealth of Nations Illustrated by academic journals: Non-profits supply most of the citations. For-profits collect most of the money.

Survey of Physics Chairs and Librarians*---Optical Society of America Which kind of journal do you think is more costly to libraries? ___Non-profit journals ___They cost about the same ___For-profit journals *Replies received from 69 librarians and 26 physics chairs.

Which costs more? Non-profits 0% About same 4% For-profits 96% Librarians say: Non-profits 0% About same 4% For-profits 96% Physics Dept Chairs say: Non-profits 8% About same 38% For-profits 53%

What do we conclude? Librarians are brighter than physics chairs? Choosers don’t know the costs of what they demand. (also true of textbooks, medicine) Scholars donating articles and referee reports to overpriced journals don’t know they are being exploited. Libraries and societies should publicize the facts of journal pricing.

Monopoly Profits in Academic Publishing? Hint: University press and professional society journals are usually not subsidized They charge 1/5 as much per page as for-profit journals.

Elsevier Financial Statement Elsevier reports revenue 2 billion Euros in 2002. Claims to have 3d biggest internet revenues, behind AOL and Amazon. Reported profits equal 33.6% of revenue.

Why “only” 33 % profit margin? If they charge 5 times as much as non-profits, why aren’t Elsevier’s profits even greater? Smaller subscription bases, due to high price Rent dissipation Elsevier CEO got ~$2 million pay and $8 million in shares this year. New stock options of ~$32 million for top execs. Armies of lobbyists

If there is free entry, how can there be monopoly? Unlike market for shoes or groceries, competitors are prevented by copyright from offering perfect substitutes. Coordination and reputation makes it hard for new entrant to attract top quality articles.

Fable of the Anarchists’ Annual meeting Once upon a time a bunch of anarchists happened to get together on January 3 in a hotel in Kansas City They had a grand time. Next year more anarchists came and they had even more fun. The tradition grew and meetings got bigger and more enjoyable.

Trouble in Kansas City One year, the hotel owner raised his rates at conference time. Attendance fell a little and owner’s revenue rose a lot. Next year owner did it again. Anarchists groused, had less fun with the smaller crowd and higher prices. Why didn’t they move to another hotel? They are anarchists!

Economic weirdness? If a Volvo cost 6-15 times as much as a Saab, how many Volvos would be sold? Almost zero, because people would substitute Saabs for Volvos. Why then do commercial journals that cost 6-15 times as much per cite as nonprofits continue to sell?

Journals as Complements Academic journals tend to be complements, not substitutes. Two copies of cheap society journal will not replace a subscription to Elsevier journal that costs 10 times as much per cite. Scientists want to read ALL significant research in their area.

New Journal Pricing Models Electronic distribution allows new pricing methods not available with paper. University-wide site licenses Price discrimination by size of university Bundling of Journals with all-or-nothing pricing Consortium pricing

Benefits of Price Discrimination and Bundling Price discrimination allows access for small colleges, poor countries. Bundling allows big universities to get everything a publisher produces. Cost of access for extra subscriber is nearly zero, so this improves efficiency.

Benefits for Whom? Profit maximizing seller can charge close to total willingness to pay of university users, leaving no net benefit to subscribers. This is worse for users than outcome if libraries refused to buy overpriced site licenses and forced seller to deal with one user at a time.

A Publisher’s View “So, we should have models where we make a deal with the university, the consortia or the whole country, where for this amount we will allow all your people to use our material, unlimited… And, basically the price then depends on a rough estimate of how useful is that product for you; and we can adjust it over time. It is a principle, which, in my view, is not immoral.” From a speech by Derk Haank, CEO, Elsevier Science

Price discrimination by non-profits With non-profits, a different story. Price discrimination & bundling improve efficiency and access. Library site licenses provide revenue to cover costs and if properly priced, allow access to nearly everyone. Probably should be priced at marginal cost (free?) to small libraries.

More strange economics With most goods, middleman pays producer, consumer pays middleman. With journals, producer pays middleman (often not much), consumer pays middleman

Follow the money… Ordinary goods Academic Journals

Open Access Model Producer pays middleman, consumer pays nobody. Would this work for nonprofits? Would this work for profit-maximizers?

Non-profit open access? To succeed an open access journal must attract authors. Are authors and their universities willing to pay to have their work read and cited? Evidence that open access articles are more cited. Economic study: avg citation worth $40 per year. But will they pay $1500 as for PLOS.

Open access and competition Competition for authors will be stiffer than for readers. Would an author submit papers to a journal with submission fees 6-15 times as high as equivalent competitor? Not likely. Why? For authors, journals are substitutes, not complements.

University-paid author fees University could limit amount it would pay per page. Authors can choose to top up fees. As outlets, competing journals are substitutes, not complements Price competition for author fees is likely to prevent exhorbitant fees.

Entry of New Journals The most interesting defense for Elsevier is that they have introduced more new journals than the non-profits Was this a good thing in the past? Is it still a good thing?

A bit of history In 1960 there were ~30 econ journals, all non-profit, all cheap. In 1980,~120 econ journals, half for-profit, half non-profit. In 2000,~300 econ journals, 2/3 for-profit. Between 1985 and 2000, top ten for-profits almost doubled their pages, top ten non-profits increased ~20%.

Entrepreneurship In the 1970’s the economics profession grew rapidly . Society journals did not grow as fast. Specialized field journals were rare. North-Holland (and others now merged into Elsevier) started about 30 new journals. About 12 of these are now leading field journals.

Rewards to innovation In 1985, the new commercial journals cost about 3 times as much per page as their non-profit counterparts. Now about 6 times as much and increasing by 6.5% per year.

Why didn’t economics societies start more journals and add more pages? Confidence that private sector would do as well as non-profit. Elitist views of old-guard leadership. Short, traditionally honorary, terms of society leaders.

Pages published by U.S. Societies Discipline Society members Pages published Microbiology 42,000 53,000 Physics-APS 40,000 96,600 Math-AMS 30,000 12,000 Optics-OSA 13,000 19,000 Econ-AEA 18,800 4,300 Entomology 6,500 4,100 Polit. Science 14,000 2,000

New Journals, Bundling, and Electronic access Commercial publishers can start new journals and add to the “Big Deal” bundle. They can monitor downloads at each university and adjust that university’s price to capture total benefit. Solves coordination of subscriptions. Who gains from existence of new journal?

Bundling and Entry Deterrence Elsevier’s bundling policy deters potential entrants. Elsevier prices rise about 7% per year. Library budgets grow less rapidly. This leaves no room in budget for new cheaper journals unless library drops entire Elsevier collection.

Non-profit startups Tough for independent new entrant. Possible for society if it is willing to bundle and price discriminate. Open access model may solve the problem for independents.

Had enough? OK, then, I’ll quit

Want more? Check out www.econ.ucsb.edu/~tedb …papers, statistics, weasel’s manual,etc

Want more? www.econ.ucsb.edu/~tedb …papers, statistics, weasel’s manual, etc

Collective Action? Acting jointly, libraries should simply refuse to buy site licenses for much above average cost. Overpriced journals would have to either cut prices or lose readers, authors, and editors.

What should scholars do? Refuse to referee for overpriced journals. Encourage cheap journals. Referee for them. Cite them. Publish in them. Encourage professional societies to expand their journals and start new ones. Keep copyright on your own work and keep all of your papers on the web.

The Cost of Going Online In 1998 almost no journals were online. In 2002, almost all were. Top non-profits $0.10/page in 1998 and $0.12 in 2002 --paper plus online. Top for-profits $0.79/page in 1998 and $0.89 in 2002 --paper plus online. Conclusion: Going online cost about $.02 per page. Prices of for-profits are not determined by costs.